Catherine Shi, partner and head of overseas business of DKL, wishes to share with you Mr. Kan’s insights on the ethical considerations for artificial intelligence with an international perspective. 

On 28 September 2018, at the conclusion of the 2nd Belt and Road Law Conference organized by The Law Society of Hong Kong, 34 law associations from 17 jurisdictions signed the following declaration on the formation of the Law Tech Alliance on promoting a shared ethical artificial intelligence (AI) framework between lawyers from the “Belt and Road” regions:

We, the undersigned organisations, share our support for innovation and technology development. We are also cognizant that there are legal and ethical challenges to the design, development and deployment of artificial intelligence around the world which may pose threats to the human race. There are many issues, such as whether A.I. should be used against human in warfare, in profiling and discrimination;whether communications and decision-making by and between machines should be fair and be examinable intelligently by human; effect of A.I. on rights and expectations to data protection and privacy; speed of replacement of traditional jobs and capacity building for new jobs, etc. They all need to be addressed. We therefore agree to collaborate to raise and critically examine relevant issues from a legal perspective, to heighten public awareness and to make submissions to our respective governments and the United Nations General Assembly. 

The 34 signatories of the declaration on the formation of the Law Tech Alliance included, in addition to The Law Society of Hong Kong, 15 Chinese Mainland law associations, the Taiwan Bar Association, the Macau Lawyers Association, the Bar of Attorneys in Gdańsk, the Beirut Bar Association, the Czech Bar Association, the Federal Chamber of Lawyers of the Russian Federation, the French National Bar Council, the Indian National Bar Association, the Luxembourg Bar Association, the Milan Bar Association, the Mongolian Bar Association, the Myanmar Bar Association, the German Federal Bar, the Law Society of Brunei Darussalam, the Law Society of Singapore etc.

It is therefore most heartening to note that at the 41st session of the General Conference of the United Nations Educational, Scientific and Cultural Organisation (UNESCO) held in Paris on 24 November 2021, 193 Member States passed a resolution to adopt the “Recommendation on the Ethics of Artificial Intelligence” (“Recommendation”), the first ever global standard on the ethics of artificial intelligence adopted by the Member States of UNESCO at the General Conference.  According to Audrey Azoulay, Director-General of UNESCO, this historical text defines the common values and principles which will guide the construction of the necessary legal infrastructure to ensure the healthy development of AI.  

Perhaps it is interesting to note that China is a Member State of UNESCO, and the United States is not, as President Donald Trump took the United States out of UNESCO in 2017.  

The preamble of the Recommendation canvasses the sweeping premise upon which the Recommendation has become pressingly important:   

1.AI has profound and dynamic positive and negative impacts on societies, environment, ecosystems and human lives, including the human mind. 

2.AI technologies can be of great service to humanity and all countries can benefit from them, but they also raise fundamental ethical concerns. AI technologies can deepen existing divides and inequalities in the world, within and between countries. Bringing benefits can also amplify tension around innovation, asymmetric access to knowledge and technologies. Justice, trust and fairness must be upheld so that no country and no one should be left behind.  

3.A standard-setting instrument developed through a global approach, based on international law, focusing on human dignity and human rights, as well as gender equality, social and economic justice and development, physical and mental well-being, diversity, interconnectedness, inclusiveness, and environmental and ecosystem protection, can guide AI technologies in a responsible direction.   

The Recommendation seeks to define AI systems as systems which have the capacity to process data and information in a way that resembles intelligent behaviour, and typically includes aspects of reasoning, learning, perception, prediction, planning or control.  

The principal aim of the Recommendation is to provide a basis to make AI systems work for the good of humanity and to prevent harm.  Its principal objective is to provide a universal framework of values, principles and actions to guide States in the formation of their legislation, policies or other instruments regarding AI, consistent with international law.       

The values propounded hinge on the respect, protection and promotion of human rights and fundamental freedoms and human dignity. Environmental and ecosystem flourishing should be recognised, protected and promoted through the life cycle of AI systems as well as ensuring diversity and inclusiveness.  Furthermore, AI actors should play a participative and enabling role to ensure peaceful and just societies, which is based on an interconnected future for the benefit of all.  

The Recommendation lays down the following fundamental principles:

1.There is to be proportionality in that AI system life cycle should not exceed what is necessary to achieve legitimate aims and objectives.  In scenarios where decisions are understood to have an impact that is irreversible or difficult to reverse or may involve life and death decisions, final human determination should apply.  In particular, AI systems should not be used for social scoring or mass surveillance purposes.  

2.Safety and security risks should be avoided.  Safe and secure AI will be enabled by the development of sustainable, privacy-protective data access frameworks that foster better training and validation of AI models utilizing quality data. 

3.AI actors should promote social justice and safeguard fairness and non-discrimination of any kind in compliance with international law. 

4.The advent of AI technologies can either benefit sustainability objectives or hinder them.  Therefore, the continuous assessment of AI technologies on sustainability should be carried out against a set of constantly evolving goals across a range of dimensions, such as those currently identified in the Sustainable Development Goals of the United Nations.  

5.There should be a right to privacy and data protection monitored by judicial systems. 

6.It must always be possible to attribute ethical and legal responsibility for any stage of the life cycle of AI systems, as well as in cases of remedy related to AI systems, to physical persons or to existing legal entities.  Life and death decisions should not be ceded to AI systems.  Human oversight must include also public oversight.  

7.The transparency and explainability of AI systems are essential preconditions to ensure the respect, protection and promotion of human rights, fundamental freedoms and ethical principles.  Transparency is necessary for relevant national and international liability regimes to work effectively. A lack of transparency could also undermine the possibility of effectively challenging decisions based on outcomes produced by AI systems and thereby infringes the right to a fair trial and effective remedy, and limits the areas in which these systems can be legally used. 

8.Appropriate oversight, impact assessment, audit and due diligence mechanisms, including whistle-blowers’ protection, should be developed to ensure accountability for AI systems and their impact throughout their life cycle.  Both technical and institutional designs should ensure auditability and traceability of the working of AI systems, particularly to address any conflicts with human rights norms and standards and threats to environmental and ecosystem well-being.  

9.Public awareness and understanding of AI technologies and the value of data should be promoted through open and accessible education, civic engagement, digital skills and AI ethics training, media and information literacy and training.

10.Participation of different stakeholders, throughout the AI system life cycle, is necessary for inclusive approach to AI governance, enabling the benefits to be shared by all and to contribute to sustainable development.

As to policy action enabling the effective implementation of the Recommendation, UNESCO will therefore:  (1) develop a readiness assessment methodology to assist interested Member States in identifying their status at specific moments of their readiness trajectory along a continuum of dimensions; and (2) ensure support for interest Member States in terms of developing a UNESCO methodology for Ethical Impact Assessment (EIA) of AI technologies, sharing of best practices, assessment guidelines and other mechanisms and analytical work.  

On the other hand, governments should adopt a regulatory framework that sets out a procedure, particularly for public authorities, to carry out ethical impact assessments on AI systems to predict consequences, mitigate risks, avoid harmful consequences, facilitate citizen participation and address social challenges.  

Member States should ensure the institution of a governance and stewardship regime.  Harms caused through AI systems are to be investigated and redressed.  Certification mechanism for AI systems and the mutual recognition of their certification are to be developed.When developing regulatory frameworks, Member States should take into account that ultimate responsibility and accountability must always lie with natural or legal persons and that AI systems should not be given legal personality themselves. 

作者简介

简家骢

DKL 香港创始合伙人